The mistakes post-production cannot fix
Six faults an edit cannot recover, each with the reason it is gone and the cost of preventing it, which is usually nothing. The honest list is shorter than the ones you have read, and the shortness is the useful part.
You have probably seen a post promising ten things post-production cannot fix. The number is doing work the items cannot. Six of them are absolute. The other four are expensive rather than impossible, which is a different problem with a different answer, and stapling the two together is how the list gets to ten.
Worth saying first: an edit is very good at the faults people panic about. Colour that swings warm then cold, a video that is quieter than everything else on the platform, a fluffed line, a fridge humming in the background, a shot that drifts out of focus for two seconds. All work, all with a cost attached, all recoverable.
What follows is the other category. Six faults, each unrecoverable for a reason you can state in a sentence, and each preventable for nothing.
What can an edit actually recover?
One test covers almost everything. Ask whether the thing you want is somewhere in the recording, buried under something else, or whether it was never recorded at all.
Anything buried is a cost. Noise sits alongside a voice and can be pulled away from it. A flat picture holds colour information that a grade brings forward. A slow section holds a fast one inside it, and cutting finds it. None of that is free, but all of it is possible, because the material exists.
Anything absent is a reshoot. One fault below is the borderline case: the material technically survives, mixed in, but the rescue is uncertain enough that it belongs here with the missing rather than with the fixable. That is the list below.
What is genuinely gone?
The room was louder than the person
If the microphone sat far enough away that the reflections off your walls arrived at close to the same strength as your voice, both were written into the same file as one signal. Strictly the voice is not gone, it is buried, and the newer tools that separate a voice from a room can sometimes pull it back. It sits on this list rather than the recoverable one because that rescue is uncertain, and pushed far enough it strips the top off the voice as well, so you swap a room for somebody who sounds like they are on a bad phone line. When the room is that loud a reshoot usually costs less than the fix and comes back better. Prevention is moving the microphone nearer, which takes ten seconds and costs nothing.
The level was too high
Clipping is not damage added to a recording. It is information missing from one. The loudest moments hit the ceiling of what the file can hold, the peaks are flattened off, and what happened up there was never written down. Turning it down later gives you a quieter copy of the same flat tops. Prevention is setting the level so that the loudest thing you do, which is usually a laugh, lands comfortably short of the top, then not touching it again.
There was nothing to cut to
A cut needs somewhere to go. When a moment exists on one angle only and something in the middle of it has to come out, the join announces itself, because the frame before and the frame after are the same frame with a piece missing. Editors have covers for this and every one of them is a compromise a viewer can see. Prevention is a second angle, or a few seconds of anything else happening in the same room, filmed at the same time so the light matches.
The highlights went white
An overexposed area is not bright. It is empty. Every pixel in it holds the same maximum value, so pulling the exposure down afterwards produces grey where the detail should be, not the window, the shirt or the lit side of your face. Prevention is exposing for the brightest thing you actually care about and letting the rest fall dark, checked once before you start rather than discovered at the end.
The close-up was never filmed
You cannot crop into a shot you did not take. Enlarging a wide frame throws away resolution, and past a certain point the image goes soft in a way that reads as a mistake rather than a choice. If the hands, the pan or the page is the lesson, it has to be filmed at the size it matters at. Prevention is one extra pass over the same action, or a phone left running on the detail while you work.
Something changed halfway through
Hair, wardrobe, the position of the lamp, the height of the chair, the state of the desk behind you. When any of those changes mid-session, the two halves stop cutting against each other and every join between them is visible. Nothing in an edit puts a jumper back on. Prevention is a photograph of your own setup, taken on your phone before you start, looked at again after every break.
Why is the honest list this short?
Because a great deal of what gets called unfixable is expensive rather than impossible, and the two get filed together by people who have only ever been on one side of the handover. Newer tools blur the edge a little, an upscale here, a rebuilt peak there, but they do it by inventing plausible detail rather than returning the real thing, so the six faults above stay absent even when the software makes them look present.
Keeping them apart is worth more than any individual item above. A fault that is expensive is a negotiation: how much does it matter, what would it cost to improve, is that a fair use of the budget. A fault that is absent is not a negotiation at all. There is no setting, no plugin and no amount of money that puts information into a file which never held it.
That distinction also tells you when a supplier is guessing. Anybody who promises to fix a blown window or unclip a peak is either misunderstanding the question or hoping you will not check.
What does this change before your next shoot?
Six checks, none of which costs anything: microphone closer, level lower, a second angle or a few seconds of cutaway, expose for what matters, film the detail at detail size, photograph your setup before you begin.
If you already have footage with one of these faults in it, the honest question is not how to rescue it. It is whether the moment is worth filming again. Often it is, because the reshoot takes an afternoon and the rescue takes a fortnight and still looks like a rescue.
Keep reading
This is a service, and the method is written down.
Everything above came out of doing the work rather than writing about it. If you want the method instead of the story, it runs in order on one page.
Start with a free audit
Tell us where your content is now. We will come back with what we would change and what result to expect.
A person reads the channel and writes the audit by hand: a considered read typically takes three working days. That is the usual shape, not a promised turnaround. We use these details only to reply to you: no lists, no lurking.
What you will get
A fit snapshot: where your channel stands, and whether we are a match.
Two to three opportunities: specific, prioritised, yours to keep.
A recommended next step, even if that step is not us.
The audit is free and commits you to nothing: nobody follows up with a call you did not ask for.